Abstract
Breast cancer accounts for approximately 20% of all female malignancies with hereditary breast cancer being implicated in 5-10% of these cases. Two highly penetrant hereditary breast cancer genes are known; BRCA1 (17q) and BRCA2 (13q), which also confer an increased risk of cancer at other sites. The risk of breast and ovarian cancer in BRCA1 mutation carriers appears to be modified by a number of factors, including parity, age of first birth and position of the BRCA1 mutation. We have shown that there is a 2-fold increased risk of ovarian cancer in BRCA1 carriers with rare alleles of the HRAS 1 VNTR versus those with common alleles only. We did not observe a difference in breast cancer risk with the presence of rare alleles.
This is one of the first examples of a gene modifying the penetrance of an inherited cancer gene (Paper I). The proportion of families attributable to BRCA2 is less than previously estimated (Papers II and III). In addition to an early age of onset of female breast cancer (less than 35 years), families with BRCA2 mutations are also at an increased risk for male breast, pancreatic, and prostate cancer. Recurrent mutations in conjunction with a common 13q haplotype have been observed in some individuals suggesting that founder effects are present in the inheritance of this disease (Paper III). LOH on chromosome 13 in sporadic breast and ovarian tumours has implicated BRCA2 as a candidate TSG. However, despite screening 70 breast and 55 ovarian tumours, we identified very few BRCA2 mutations. This lack of significant mutation in BRCA2 (and BRCA1) in sporadic breast and ovarian tumours suggests that either different mechanisms of tumour suppressor gene inactivation play a role or different genes are involved in the sporadic versus familial cases (Paper IV). The MDGI gene (lp33-p35) has previously been implicated as a breast cancer TSG. However, we have not identified inactivating mutations in the coding region in a panel of sporadic breast tumours (Paper V).
More recently, it was shown that the MDGI gene expression is shut-off by aberrant methylation in breast carcinomas and cell lines. Thus the MDGI gene may not be the target of LOH on chromosome lp in breast cancer. In a study of 1280 breast carcinomas, we found multiple regions of LOH on chromosome 17 involved in human breast cancer. The regions were defined by multiple independent deletions and by statistically significant associations between LOH and specific clinicopathological features. These include age of onset, family history of breast cancer, tumour histopathology, tumour size, estrogen receptor (ER) status and occurrence of Iymph node or distant metastases. Each of the clinical parameters except distant metastases was significantly associated with loss in specific subregions with ER-negative and ductal tumours showing LOH for markers along the majority of the chromosome (Paper VI).
